Home / OSS Directory
OSS Directory
Open source tools for red teaming, agent and tool security, runtime guardrails, model and artefact supply chain, evaluation, privacy and interpretability, assessed for what they actually cover and where they fit in a programme.
| Record | Purpose | Maintainer | Licence | Stars |
|---|---|---|---|---|
| 0A cybersecurity harness for full-stack LLM-driven penetration testing. Find and fix vulnerabilities autonomously, 24/7. [RESEARCH PREVIEW] | MCP server scanning | 0sec-labs | Custom or multiple | 80 |
| 0day-RubbishRedefining vulnerability disclosure in the AI era. We mass-produce exploitable 0days and disclose them directly, using event-driven pressure to… | Red teaming automation | Exploit-Garbage | MIT | 218 |
| a2a-demosDemo agents showcasing CapiscIO Agent Guard and MCP Guard — trust badges, identity verification, and tool-level authorization for A2A and MCP… | MCP server scanning | capiscio | MIT | 23 |
| ADRUber's open-sourced agent detection and response framework for enterprise AI agents, combining runtime observability of agent actions with… | Agent runtime firewalls | uber | Apache 2.0 | 1,583 |
| AdrianOpen-source runtime AI agent security tool - monitors and controls AI agents, catching malicious tool use, prompt injection, and policy drift in… | Agent runtime firewalls | secureagentics | Apache 2.0 | 573 |
| AdStrikeAI-powered modular Active Directory red-team framework for authorized penetration testing, AD enumeration, attack-path analysis, Kerberos/ADCS… | Red teaming automation | capture0x | MIT | 354 |
| AegisLocal, monitor-first observability for AI agents: processes, file activity, TCP endpoints and attribution evidence. Windows primary; macOS/Linux… | Drift and performance monitoring | Acacian | MIT | 16 |
| aegisoraThe Zero-Trust Runtime Security & Governance Layer for Autonomous AI Agents. | MCP server scanning | aegisora-ai | Custom or multiple | 22 |
| agent-aegisAgentAegis is a lightweight plugin providing full-lifecycle runtime protection for OpenClaw. | Robustness and stress testing | antgroup | Apache 2.0 | 199 |
| agent-airlockA type-checker and contract layer for AI agent tool calls, deny-by-default, in-process, Pydantic-only. Strict argument validation, ghost-argument… | MCP server scanning | sattyamjjain | Apache 2.0 | 15 |
| agent-auditStatic security scanner for LLM agents — prompt injection, MCP config auditing, taint analysis. 51 rules mapped to OWASP Agentic Top 10 (2026).… | Prompt injection detection | HeadyZhang | MIT | 232 |
| agent-beaconAgent Beacon is the world's first open-source telemetry layer for AI agents wherever they run: locally, in CI, in the browser, or in the cloud. | Drift and performance monitoring | Asymptote-Labs | MIT | 458 |
| agent-bomOpen security scanner and self-hosted control plane for AI, MCP, and cloud. One evidence model — run scans in your environment, centralize… | MCP server scanning | msaad00 | Apache 2.0 | 31 |
| AgentCapture针对 AI 自动化渗透 Agent 的新一代反制蜜罐，通过反向代理将API密饵载入真实业务、反向提示词注入等方式反制自动化渗透 Agent，实现多款主流通用Agent的反制上线控制。 | Robustness and stress testing | Tcotl | AGPL 3.0 | 109 |
| agent-chaperoneScreens an AI agent's tool calls before they run and tool results before the agent reads them. An MCP proxy plus a hooks adapter for a client's… | MCP server scanning | agent-chaperone | Apache 2.0 | 20 |
| agent-controlCentralized agent control plane for governing runtime agent behavior at scale. Configurable, extensible, and production-ready. | Input and output guardrails | agentcontrol | Apache 2.0 | 308 |
| agent-egress-benchOpen Apache-2.0 corpus, runner, scoring, and result-verification contracts for measuring AI-agent egress controls. A PipeLab open project. | MCP server scanning | luckyPipewrench | Apache 2.0 | 22 |
| Agent-Execution-PartnershipAgent Execution Partnership AEE is an open-source control plane that ensures every AI agent action is authorized before it runs, observable while… | Drift and performance monitoring | eli-labz | Custom or multiple | 153 |
| agentggOpen source agentic SAST. The engine behind hundreds of disclosed zero-days. 100+ AI security agents, any repo or PR diff, bring your own model. | Red teaming automation | agentgg-dev | Apache 2.0 | 300 |
| Agent Governance ToolkitRuntime governance framework for autonomous agents with deterministic policy enforcement on tool calls, zero-trust agent identity, execution… | Agent runtime firewalls | Microsoft | MIT | 6,251 |
| AgentHoundOffensive security framework for AI agent infrastructure - recon, credential looting, model exfiltration, poisoning, and attack-path analysis… | Red teaming automation | Telefonica | AGPL 3.0 | 16 |
| Agentic-Bug-HunterAI-powered bug bounty hunting toolkit that works with or without subscription. | Red teaming automation | awarexone | MIT | 5,144 |
| agentic-osGovernance framework for AI coding agents. It runs them through a five-step workflow (plan, build, review, test, ship) where no step counts as… | Input and output guardrails | KbWen | MIT | 178 |
| AgentLensAgentlens is a trusted agent trading platform. Here, you can quickly find the Agent that meets your needs, and you can also publish your own Agent… | Model signing and attestation | ZhangJinHaHaHa | AGPL 3.0 | 1,031 |
| agent-opforOpen-source adversary emulation for AI agents and MCP servers. | Jailbreak and attack libraries | KeyValueSoftwareSystems | Custom or multiple | 580 |
| agent-safe-pipelineReference architecture for AI agents that propose actions but cannot authorize them — immutable intent capture, an independent Decionis policy… | Red teaming automation | decionis | Apache 2.0 | 532 |
| agent-safety-eval-labAgent trace and tool-use safety evaluation lab. | Red teaming automation | YutoTerashima | MIT | 309 |
| AgentScanAI protocol scanner — discover exposed MCP servers, A2A Agent Cards, and open LLM APIs in one command. 一条命令盘点 MCP、A2A Agent Card 和 LLM 开放接口暴露面。 | Robustness and stress testing | 7anX | MIT | 110 |
| agentsealSecurity toolkit for AI agents. Scan your machine for dangerous skills and MCP configs, monitor for supply chain attacks, test prompt injection… | MCP server scanning | getagentseal | Custom or multiple | 374 |
| agentsecretsZero-knowledge credentials infrastructure built for AI agents to operate, not just consume. | Red teaming automation | The-17 | MIT | 180 |
| Agent-Security-Regression-HarnessExecutable security regression testing for agentic applications and MCP-integrated systems. | MCP server scanning | OWASP | Apache 2.0 | 51 |
| agent-security-scanner-mcpSecurity scanner MCP server for AI coding agents. Prompt injection firewall, package hallucination detection (4.3M+ packages), 1000+ vulnerability… | Prompt injection detection | sinewaveai | MIT | 122 |
| agentshieldAI agent security scanner. Detect vulnerabilities in agent configurations, MCP servers, and tool permissions. Available as CLI, GitHub Action, ECC… | Agent runtime firewalls | aiconnai | Custom or multiple | 19 |
| agent-shieldMulti-engine security scanner for AI agents, MCP servers & plugins — 13 engines, one report. | MCP server scanning | elliotllliu | MIT | 15 |
| agent-skills-hubDiscover and compare open-source Agent Skills, tools & MCP servers — with quality scoring, trending analysis, and automated GitHub sync | Robustness and stress testing | zhuyansen | Custom or multiple | 389 |
| agents-mdAGENTS.md patterns for context engineering in coding agents: safer command output, token efficiency, validation, and prompt-injection resistance. | Robustness and stress testing | Austin1serb | Custom or multiple | 153 |
| agent-threat-rulesOpen detection-rule standard for AI agent security threats — like Sigma, but for AI agents. Executable rules across 10 categories; merged into… | Agent runtime firewalls | Agent-Threat-Rule | MIT | 399 |
| AgrusScannerWindows network scanner that finds AI/ML services and MCP servers on your network (shadow AI discovery). Signed, self-updating detection… | MCP server scanning | NYBaywatch | Custom or multiple | 25 |
| aibeatBreak your AI before they do. Join Discord: https://discord.gg/8A6mFckxZ | Robustness and stress testing | tophant-ai | Custom or multiple | 1,063 |
| aibomAI Bill of Materials through source code scanning | AI bill of materials | cisco-ai-defense | Apache 2.0 | 109 |
| ai-bomAI Bill of Materials — discover every AI agent, model, and API in your infrastructure | AI bill of materials | Trusera | Apache 2.0 | 334 |
| ai-dfir-toolkitA vendor-neutral collection of Sigma, YARA, and Suricata rules for detecting compromise of LLM applications, MCP servers, ML supply chains, AI… | MCP server scanning | depalmar | Apache 2.0 | 23 |
| ai-driven-developmentPractices, protocols, and skills for AI-driven software development. Skills and safety hooks for Claude Code, Codex, OpenCode, Cursor,… | Robustness and stress testing | CodeAlive-AI | MIT | 138 |
| ai-engineer-notebooksHands-on, framework-free Colab notebooks for the AI Engineer / Forward Deployed Engineer (FDE) skill set — model APIs, structured output, tool… | Input and output guardrails | calmrocks | MIT | 636 |
| ai-gatewayUnified AI Gateway for 30+ LLMs (OpenAI, Anthropic, Bedrock, Azure etc) with Caching, Guardrails, A/B test & cost controls. Go-native Fastest &… | Input and output guardrails | ferro-labs | Apache 2.0 | 256 |
| aigisGet Claude Code and other AI agents approved for use at work — deterministic guardrails on every tool call, tamper-evident audit logs, and a… | MCP server scanning | killertcell428 | Apache 2.0 | 54 |
| AIHoundCredential Scanner for Popular Desktop AI Platforms | MCP server scanning | netwrix | MIT | 16 |
| ai_osint烙 Curated AI OSINT resources — Google dorks, Shodan queries, GitHub dorks, and techniques to discover exposed LLM endpoints, leaked AI API keys,… | MCP server scanning | 7WaySecurity | Custom or multiple | 166 |